original article

The Repertory Grid Technique as a Method for the Study

of Cultural Differences
Oscar Tomico 1,*, Evangelos Karapanos 1, Pierre Lvy 1, Nanami Mizutani 2, and
Toshimasa Yamanaka 2
1
2

Eindhoven University of Technology, Eindhoven, the Netherlands

University of Tsukuba, Tsukuba, Japan

Culture is typically approached in the field of design through generic, cross-domain constructs. In this paper we provide an alternative
methodological approach to exploring cross-cultural differences by studying the idiosyncratic views of individuals with regard to existing
products. We operationalize this approach through the Repertory Grid Technique, a structured interview technique motivated by Kellys
Personal Construct Theory, and propose a content-analytic procedure combining quantitative and qualitative information. We further propose
the use of three distinct metrics in the analysis of personal constructs: dominance, importance, and descriptive richness. Dominance of a
construct is measured through the relative percentage of a construct category over the total sample of constructs. Importance is measured
through the elicitation order; this assumes that constructs elicited first are more salient and important to the individual. Descriptive
richness relates to the diversity of a class of constructs. Some constructs might be uni-dimensional while others might tap to a number
of distinct facets. The use of these indices enables the quantification of the different ways in which individuals perceive and differentiate
between products. By identifying how individuals respond to a rich set of stimuli within a given domain, we inquire into their values and
the qualities they appreciate within this restricted domain. Cultural values are thus explored in relation to a set of stimuli. We tested this
procedure through an exploration of the ways 17 Dutch and 16 Japanese industrial designers valued a set of pens.
Keywords Repertory Grid, Cultural Differences, Designers Perceptions, Product Attribute Prioritization Measurements.
Relevance to Design Practice This study illustrates how the Repertory Grid Technique can be used to determine differences in product
attribute prioritization.
Citation: Tomico, O., Karapanos, E., Lvy, P., Mizutani, N., & Yamanaka, T. (2009). The repertory grid technique as a method for the study of cultural differences. International
Journal of Design, 3(3), 55-63.

Introduction

These approaches typically involve gathering information

about cultural differences through questionnaires or interviews
using previously validated and standardized items. One could,
however, wonder about the fruitfulness of applying cross-domain
cultural dimensions to domain-specific design choices. While an
overall differentiation of two groups on a given cross-domain
cultural dimension (e.g., the need for autonomy) might have
certain design implications for the design of a given product,
there might be plenty of other aspects that differentiate these two
groups within a specific context.
In this paper we propose a subjective approach to the
exploration of culture in product design, based on Kellys (1955)
theory of Personal Constructs. This approach takes individuals
perceptions of products to be a carrier of implicit cultural insight,

With the emergence of the global marketplace, cross-cultural

differences are becoming increasingly recognized as a key factor
in the successful adoption of new products (Lee & Harada, 2000).
It is thus not surprising that designers have begun to consider the
role of culture in design and to develop methods and processes for
taking cultural aspects into account when designing new products.
Culture, according to Kluckhohn (1951), is rooted in the
values that pervade the historically derived ideas that form a
particular tradition. These ideas and values create patterned ways
of thinking, feeling and reacting, which constitute the distinctive
character of a human group. Culture, however, has been largely
approached in the design field through generic, cross-domain
constructs such as that of Hofstedes (1984) cultural dimensions
and Scwhartzs (1992) cultural values (e.g., Marcus, 2000;
Oshlyansky, Cairns, & Thimbleby, 2006; Tong & Robertson,
2008). Schwartz proposed that a taxonomy of seven distinct
cultural value types can serve to distinguish different cultural
groups. These value types have been found to be universal across
cultures, but the relative dominance of each value type differs
across different cultural groups. These value types can thus be
used to predict attitudes towards external stimuli or behaviors.

www.ijdesign.org

Received April 1, 2008; Accepted September 28, 2009; Published December 28,
2009.
Copyright: 2009 Tomico, Karapanos, Lvy, Mizutani, and Yamanaka. Copyright
for this article is retained by the authors, with first publication rights granted to the
International Journal of Design. All journal content, except where otherwise noted,
is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5
License. By virtue of their appearance in this open-access journal, articles are free
to use, with proper attribution, in educational and other non-commercial settings.
*Corresponding Author: o.tomico@tue.nl

55

International Journal of Design Vol.3 No.3 2009

The Repertory Grid Technique as a Method for the Study of Cultural Differences

differences. This procedure allows for a quantitative analysis

of the existing data from different grids without loosing the
individuality of the results.
The presented approach was applied to a case study that
examined how Japanese pens were perceived by Japanese and
Dutch designers. Its aim was to explore cross-cultural differences
between the product attribute prioritization of the Japanese and
Dutch designers. While a wealth of studies has explored how
culture influences the way users perceive and interact with
products (e.g., Lee & Harada 2000; Hsu, Chuang, & Chang, 2000;
de Leur, Drukker, Christiaans, & de Rijk, 2006), we are not aware
of any efforts to explore whether designers from diverse cultural
groups differ in the ways they perceive products and prioritize
design attributes. Such cultural differences would evidently
impact decision-making in the design process and therefore the
design outcome.
In the following, we elaborate on the theory of Personal
Constructs and the Repertory Grid Technique as a methodological
approach to exploring the ways that designers perceive
products and prioritize design attributes, and we illustrate an
operationalization of this approach using the findings of the case
study.

and thus, cross-cultural differences can be explored within a

specific context through existing products in the market. The main
difference between an objective approach (in which knowledge is
seen as a representation of the real world and as something that
can be extracted through a hypothetic-deductive process) and a
subjective approach (in which knowledge is based on the subjects
experiences and actions and can be extracted through an inductive
process) to gathering information is the shift from validated items
to a validated procedure.
The Repertory Grid Technique was developed by George
Kelly (1955) as an application of the theory of Personal Constructs.
Since then, it has been widely applied in psychotherapy, and in
gathering information about consumer responses, and lately, it has
become increasingly popular in the HCI field (e.g., Hassenzahl
& Wessler 2000; Tomico, 2007). The Repertory Grid Technique
is a semi-structured interview technique that aims at exploring
how individuals construct the world around them. In the HCI
field, the Repertory Grid Technique is used to explore the ways
in which individuals perceive and differentiate between products.
Moreover, due to its hybrid qualitative-quantitative nature, it
allows for a wide kind of statistical analyses.
For the reasons given above, the Repertory Grid Technique
and three measurements to determine differences in product
attribute prioritization (dominance, importance and descriptive
richness) were used in this study to analyze cross-cultural

Measuring Dominance, Importance and

Descriptive Richness for Cross-cultural
Analysis

Oscar Tomico is an assistant professor in the Designing Quality in Interaction

Group of the Faculty of Industrial Design of Eindhoven University of
Technology. He received his PhD in June 2007 from the Technical University of
Catalonia ondeveloping subjective psychological exploration techniques based
on the constructivist paradigm for informational and inspirational purposes. His
research focuses on exploring relations between physical, social and virtual
domains as information interfaces for creating and sharing knowledge. His
current projects focus on the social repercussions of using intelligent systems as
information platforms in home, shop and library contexts.

From a constructivist approach (Kelly, 1955), an individuals

internal value system mediates that individuals evaluation of
external stimuli. For instance, when we meet a new person, we
might form a bipolar construct of friendly-distant in evaluating
that persons character. This construct is an evaluative judgment
about an attribute (i.e., friendliness) that is of importance to the
specific individual that formed the construct. Thus, the bi-polar
constructs that an individual forms in differentiating between
stimuli can provide a rich understanding not only of the stimuli
but also of that individuals internal value system.
Kelly (1955) proposed the Repertory Grid Technique
(RGT) as a means for eliciting the personal constructs of
individuals for a specific set of stimuli (cf. Hassenzahl &
Wessler, 2000; Karapanos, Martens, & Hassenzahl, 2009). The
technique consists of a structured interview technique based on
triading. During the triading, the stimuli products are presented
to the subject in sets of three and the subject is asked to think
of a property or quality that makes two of the products alike and
discriminates them from the third. This results in a similaritydissimilarity judgment in which the subject is asked to identify
an aspect that groups two of the products (similarity judgment)
and an aspect that differentiates them from the third (dissimilarity
judgment). By forming different triads (sets of three) out of the
larger pool of products and repeating the process, the subject will
elicit a number of bipolar constructs (pairs of opposite attributes).
The bipolar constructs that appear to dominate for this specific
individual and this specific set of products are a Repertory Grid.
Each Repertory Grid is personal and varies for each
participant in topic and number. The generated Repertory Grids

Evangelos Karapanos is a PhD candidate in the User-Centered Engineering

Group at the Faculty of Industrial Design of Eindhoven University of
Technology. He received a masters of science in Human-Computer Interaction
from University College London. His current work concerns methods and tools
for quantifying the diversity of subjective judgments of product quality. His
research interests include psychometric scales, personal attribute judgments, and
methodological aspects of longitudinal studies in the field of user experience.
Pierre Lvy is an assistant professor in the Designing Quality in Interaction
Group of the Faculty of Industrial Design of Eindhoven University of Technology.
A mechanical engineer from Compigne University of Technology, he earned
his PhD in Kansei Science from the Graduate School of Comprehensive Human
Sciences at the University of Tsukuba, Japan. His current work focuses on
sharing and its effect on creativity. His research areas of interest include Kansei
science, Kansei design, sharing processes, and Occidental/Oriental philosophical
differences applied to design.
Nanami Mizutani is a doctoral student in Kansei Cognitive Brain Science at
the Graduate School of Comprehensive Human Sciences at the University of
Tsukuba, Japan. She earned her masters degree in Kansei Science from the
University of Tsukuba, where she also previously studied industrial design. She
is currently working at the Sensory and Cognitive Food Science Laboratory of
the National Food Research Institute in Tsukuba as an internship student. Her
current research focuses on the effects of package design on taste perceptions.
She also researches users evaluations of product design using psychological and
physiological methods.
Toshimasa Yamanaka is a professor of Kansei Information in the Graduate
School of Comprehensive Human Sciences and at the Institute of Art and
Design at the University of Tsukuba. He teaches undergraduate and graduate
courses in Product Design and Kansei Information. He earned his PhD in Kansei
Science from the Graduate School of Comprehensive Human Sciences at the
University of Tsukuba. His research areas of interest include design process,
design analysis, and information design.

www.ijdesign.org

56

International Journal of Design Vol.3 No.3 2009

O. Tomico, E. Karapanos, P. Lvy, N. Mizutani, and T. Yamanaka

cross-cultural analysis as they allow the merging of information

from different grids without loosing idiosyncratic views.

are used to evaluate the products. Each participants Repertory

Grid is his or her personal semantic differential questionnaire
and can be used to rate the products. This technique makes the
information obtained more reliable and specific, as each Repertory
Grid belongs to a different individual, but it makes the comparison
between participants (i.e., between different Repertory Grids)
more difficult.
In order to overcome the idiosyncratic nature of the results
and to create a standardized classification scheme, one possible
solution is to apply content analysis (Krippendorff, 2004).
Classification schemes can be derived directly from the raw data,
i.e., the personal constructs, as described in the conventional
content analysis approach of Hsieh and Shannon (2005), and can
be combined with existing knowledge if a certain theory exists in
the respective field of study.
After applying a classification scheme to the personal
constructs, one may apply different indices in order to compare
the categories of constructs. In this article we opted to analyze
the resulting categories of constructs according to three distinct
metricsdominance, importance and descriptive richness.
The relative percentage for a given construct displays how
dominant that construct is for a group of individuals when they
are differentiating among a set of products. If, for example, one
group of participants, when asked to differentiate the products,
employs the construct ease of use significantly more frequently
than a second group of participants, one could conclude that the
first group is more concerned with how easy it is to use this class
of products.
The second criterion, which can be used for characterizing
the importance of a personal construct for a specific group, is its
elicitation order (Tomico, 2007; Feixas & Cornejo-Alvarez, 2002).
The elicitation order has been previously found in Constructivist
Psychology to measure the subjective importance of a construct
(i.e., its salience) as well as its implication potential (i.e., its
importance in preference judgments) (Mcdonagh & AdamsWebber, 1987). The elicitation order is measured through the
normalized order in which a construct appears (with constructs
reported first being considered more important to the individual
than those reported later), and the overall dominance of a construct
category may be computed across individuals as the average of
the elicitation order of the constructs from the same category.
A third criterion is descriptive richness, which can be used
to determine the reach of each category. It can be defined as the
range of different personal constructs (attributes) elicited within
the same category. The different ways in which participants refer
to the same categories relate to how the personal constructs elicited
are related to each other and how big the clusters of constructs
are. For instance, a construct category such as novelty might
have a single facet relating to the novelty and innovativeness of a
product, while a different construct, such as ease-of-use, might
tap to more than one facet, for example understandability, clarity
and navigability.
One may thus conduct analyses through the decomposition
of repertory grids into: dominance (relative percentage),
importance (elicitation order) and descriptive richness of each
construct category. These indices can provide valid information for

www.ijdesign.org

Exploring Cross-cultural Differences

between Dutch and Japanese Designers
Designers constantly make decisions. Most of these decisions are
unlikely to be grounded in empirical data. Karapanos and Martens
(2007) showed how differences in the professional backgrounds
of designers on a small design team impacted the prioritization of
design goals. In the same vein, a designers cultural background
is likely to influence the product qualities that will dominate in
the design process.
This study aimed to explore cross-cultural differences
among Dutch and Japanese designers perceptions of a set of pens.
By studying the way these designers perceived and differentiated
among a set of products (measured by the metrics of importance,
dominance and descriptive richness with regard to the different
categories generated), we aimed to understand the sets of attributes
they value, and how these values might relate to their respective
cultural backgrounds and might motivate their design decisions.

Experiment Set-up
A total of thirty-three individuals (16 Japanese and 17 Dutch)
participated in the study. Their ages varied from 23 to 32 years
(mean age for the Japanese group was 26.8, for the Dutch group
24.8), three of the Japanese and four of the Dutch participants
were female. They were all trained industrial designers, either
students of industrial design (accounting for 69.2% of the
Japanese, 88.2% of the Dutch) or researchers in the field of design
(30.8% of the Japanese, 11.8% of the Dutch) from universities
with similar approaches to design practice. Participants from both
groups had similar preferences with regard to writing tools. The
Dutch participants often used for writing (in order of importance)
ballpoint pens, keyboards, fine liners, markers and cell phones.
The Japanese participants often used (in order of importance)
ballpoint pens, keyboards, mechanical pencils, multifunctional
pens, fine liners, drawing tablets and cell phones. The average
amount of money paid for a pen by the Japanese participants was
3.6 euros (479 yen) and for the Dutch participants was 5.18 euros
(689 yen), with a maximum of 15 euros (2,000 yen) for both the
Japanese and Dutch participants.
Six different pens that were designed for the Japanese
market were employed as the stimuli products in the study (see
Figure 1). The six products used for testing were determined in a
previous study with Japanese participants to be the most preferred
out of a larger set of 20 products. The objective parameters of the
pens are described in Table 1.

Procedure
First, the six products were combined in randomly selected triads,
out of a total pool of 15 possible triads. The order in which the
three products were presented was randomized. For every triad,
participants were asked to think of a property or quality that
makes two of the products alike and which discriminates these

57

International Journal of Design Vol.3 No.3 2009

The Repertory Grid Technique as a Method for the Study of Cultural Differences

two from the third (i.e, an opposite property or quality). From

the first answer, laddering down and up procedures were applied
to the positive and negatives poles of each construct to get to the

core of the answer. The same procedure was repeated until a point
was reached at which no new attributes arose for two consecutive
triads. Finally, the subjects were asked to rate all of the products

Figure 1. Pens A-M-E (upper row) and R-C-Q (bottom row). Pen A has a transparent plastic cover, pen C and E have
semitransparent plastic cover. Pen R has a metallic cover with a matte finish. Pen Q has a metallic cover with a glossy finish.
Table 1. Objective parameters of the six pens used in the study.
A

Length

138.8

146.9

142.8

138.5

142.4

130.8

Weight

25

14

25

13

21

11

Maximum diameter

13.7

14.6

10.3

13.1

9.6

Type of ink

Water-based
biopolymer

Oil-based

Water-based
pigment

Oil-based

Oil-based

Oil-based

Color of pen ink/

size of pencil lead

black, red ink

black, red ink/,

0.5 pencil lead

black, red, blue

ink

black, red ink/ 0.5

pencil lead

black, red, blue

ink

black, red ink/ 0.5

pencil lead

None

Soft rubber

Hard rubber

None

Hard rubber

None

0.4 mm

0.7 mm

0.5 mm

0.7 mm

0.7 mm

0.7 mm

Pilot

Uni

Zebra

Zebra

Zebra

Zebra

Vertical click

Vertical click

Vertical click

Horizontal turning

Vertical click

Horizontal turning

Grip
Ball point size
Maker
Interaction
mechanism

Table 2. Sample from the repertory grid for one of the Dutch participants.
Negative Pole

Positive Pole

Underneath or below the ideal diameter (1 cm),

difficult to handle

Pens are thin for holding, easier to control, more precise

Feels like nothing is in your hand; too light

Relatively heavy; weight resistant

Too light of a plastic pen; no quality

Material has quality; robust, solid

Unattractive patterns, cheap colors; resembling a

toy;glittery

Appearance, coloring have professional resemblance; metal,

aluminum look

Crappy; lot of force needed; not working properly;

weak mechanism

Smooth, proper feedback; clicks properly; quality mechanism

Only one function, one type of writing; too small to

hold properly

Multi-functional, with an eraser. All together, 4 separate things;

refillable.

Common mechanisms that I see often

Unique switching mechanism

Gel grip, feels too soft, slippery; feels like Im loosing

the grip or like its too hard

Rigid grip, soft material feeling but not loose

Feels like you are going over a rough surface; need

to apply more force.

It has a fine liner; I can also draw with it, can make clear lines
that are smooth, soft

www.ijdesign.org

58

International Journal of Design Vol.3 No.3 2009

O. Tomico, E. Karapanos, P. Lvy, N. Mizutani, and T. Yamanaka

descriptive richness). This classification was done by the first two

authors, each working independently. The interrater agreement
(Fleiss, Levin, & Paik, 2003) of the classification process was
determined to be satisfactory (k = 0.82).
Table 3 illustrates the detailed coding scheme together with
the breakdown into thematic categories (pragmatic and hedonic),
the defining aspects of these categories, and examples of the
personal constructs elicited. As evident in the table, the pragmatic
aspects were related to utility (i.e., the quality of outcome in
using the pen), durability (i.e., the feeling of reliability), easeof-use (i.e., how easy it was to learn to operate the pen) and the
level of comfort experienced while writing with the pen. As for
hedonic aspects, stimulation was found to be induced by three
distinct aspects of the products: unexpected functionality, the
aesthetics of interaction (i.e., the degree to which the user related
to the expressiveness, richness, and therefore pleasantness of
the interactive experience) and the aesthetics of appearance, or
visual aesthetics (i.e., the degree to which the user related to the
products form and color). Finally, it was found that constructs
relating to identification were concerned with the self-image that
the pen might potentially communicate about the owner.
This hierarchical classification of the designers personal
constructs enables a decomposition of their perceptual space
into semantically distinct and culture-independent dimensions.
Next, we explored whether any cross-cultural differences existed

according to their personal attributes using a scale similar to a

semantic differential scale. See Table 2 for a sample of the results
for one participant.
A total of 190 attributes were obtained for the group of
Dutch designers and 201 for the group of Japanese designers
(with 7 to 24 attributes per participant). The interviews with the
Japanese participants were conducted in Japanese and the results
were translated afterwards into English by the interviewers. The
interviews of the Dutch participants were able to be conducted
in English as all of them had a high level of command of the
language. Personal constructs from both cultural groups were first
submitted to qualitative content analysis (Hsieh & Shannon, 2005;
Krippendorff, 2004). A detailed coding scheme emerged from
the data (see Table 3). The elicited categories were then grouped
into three overall categories that reflected Hassenzahls (2004)
distinction between pragmatic and hedonic product qualities.
Pragmatic qualities refer to the instrumental aspects of a product,
such as its usefulness, practicality and ease of use. Hedonic
aspects relate to the more experiential aspects of product use
and consist of two distinct categories: stimulation, which refers
to a products ability to address the human need for novelty and
challenge, and identification, which refers to a products ability
to address the need for expressing ones self through the objects
one owns. Next, each personal construct was then classified into
one of the three construct categories (dominance, importance, and

Table 3. Personal constructs broken down into thematic categories, with examples from both the Japanese and Dutch designers.
Category
Pragmatic

Stimulation

Examples

Utility

water-based ink; can write clearly and is easy to read

Durability

the construction feels sturdy; it feels reliable

Ease-of-use

Separate click system; the function makes sense

Comfort

soft grip; you can write without hurting your fingers

Unexpected functionality

Looks like a normal pen but it is three in one!

Aesthetics of interaction

It changes from a rotating movement to a lateral movement;

it is newer and interesting; you can play with it

Visual aesthetics

limited amount of colors; it looks more stylish

Identification

an exclusive design; it works by turning

Table 4. Personal construct thematic categories: The dominance of each category (measured by the relative percentage) and
the importance of each category (measured by the elicitation order) for the Dutch and Japanese designers. Standard deviations are
displayed in brackets.
Category
Pragmatic aspects

Dominance (relative percentage %)

Importance (elicitation order)

Dutch

Japanese

Dutch

Japanese

58

69

0.47 (0.32)

0.5 (0.33)

Utility

13

16

0.57 (0.35)

0.55 (0.32)

Durability

11

12

0.4 (0.3)

0.62 (0.33)

Ease-of-use

14

0.53 (0.33)

0.45 (0.25)

Comfort

20

34

0.39 (0.29)

0.45 (0.34)

32

28

0.54 (0.3)

0.64 (0.34)

0.55 (0.32)

0.86 (0.55)

Stimulation
Unexpected functionality
Aesthetics of interaction

0.53 (0.29)

0.69 (0.33)

Visual aesthetics

15

23

0.55 (0.31)

0.62 (0.33)

0.57 (0.31)

0.72 (0.33)

Identification

www.ijdesign.org

59

International Journal of Design Vol.3 No.3 2009

The Repertory Grid Technique as a Method for the Study of Cultural Differences

from the experience of surprise and the aesthetics of interaction.

Overall, the Dutch designers seemed to be more concerned with
early experiences, whereas the Japanese designers were more
concerned with prolonged interaction.
These findings are enforced if one looks also at the
elicitation order. Both the Dutch and the Japanese designers
tended to elicit pragmatic aspects earlier than hedonic aspects
(stimulation and identification). However, more particularly,
the Japanese designers seemed to elicit constructs related to
identification only among the very last of their constructs. This
could imply that the identification constructs that were elicited by
the Japanese designers are the result of extended use. This reduces
even more the prominence of social (i.e., identification) influences
in the Japanese designers perceptions of the product. Comfort
(i.e., long-term usability) seemed to be the most prominent
construct for both the Japanese and Dutch designers, whereas
durability appeared to be more prominent for the Dutch designers.
Constructs related to the products ability to stimulate the user
appeared to be more prominent for the Dutch designers than for
the Japanese designers.
Overall, the Dutch designers appeared to be more
concerned with social (i.e., identification) considerations of
product use, whereas the Japanese designers seemed to place
greater importance on the pragmatic aspects (i.e., utility and longterm usability) of the product.

between the Dutch and Japanese designers perceptions through

the calculation of a) the relative percentage (dominance), b) the
average order elicitation (importance) and c) the descriptive
richness of each construct category for each cultural group. It
should be noted that the order elicitation index is the average of
the order of all constructs within a construct category and that
descriptive richness relates to the different characteristics that
define a product category.

Dominance and Importance Measures

Table 4 presents the relative percentages and the average elicitation
order for each construct category for both the Dutch and the
Japanese groups. A single elicitation order index was derived for
each construct category, representing the average order (mean) of
the constructs within this category. The order of the constructs
was normalized for each participant to a range of 0 to 1 based on
the total number of constructs generated (1). A 0 value reflects
the first construct that was elicited and an index of 1 reflects the
last construct. The standard deviations are included in brackets
and are crucial for deriving an estimation of how homogeneous
a category of constructs is in the relative order with which its
constructs are elicited.

order(normalized) =

order 1
(1)
totalconstructs 1

It should be noted that pragmatic qualities constituted

the designers most frequent concerns among both the Dutch
and the Japanese participants. This resembles findings from the
Technology Acceptance literature (Venkatesh, Morris, Davis, &
Davis, 2003) and the User Experience literature (Hassenzahl,
2004; Karapanos, Hassenzahl, & Martens, 2008; Karapanos,
Martens, et al., 2009), in which the usefulness of a product
appears to be the most significant predictor of the users overall
attitude towards the product.
However, the constructs of the Dutch and Japanese
designers differed in the dominance of pragmatic and hedonic
qualities. While both groups acknowledged pragmatic quality as
the dominant quality they looked for in a pen, the Dutch designers
gave more weight than the Japanese designers to the hedonic
aspects of the product. The Japanese designers seemed to give
most importance to pragmatic aspects. They were less concerned
than the Dutch designers with the stimulation that the user might
derive while using the product. They also valued less the products
role in communicating a desired self-identity for the owner to
relevant others (i.e., identification).
More interestingly, while the Dutch designers seemed to
associate pragmatic quality with ease-of-use (i.e., learnability)
relatively more than the Japanese designers, the Japanese
designers seemed to give more weight to comfort (i.e., longterm usability) and to the products utility as the major aspects
of pragmatic quality. One could argue that the Dutch designers
were empathizing with early-use situations, whereas the Japanese
designers were more concerned with prolonged use. In the same
vein, while the Japanese group mostly associated stimulation
with aesthetic appreciation derived from visual appearance, the
Dutch designers were also concerned with stimulation derived

www.ijdesign.org

Descriptive Richness
In a final analysis, we aimed for a deeper understanding of the
ways in which the two cultural groups of designers referred to
product qualities (i.e., the descriptive richness of a category).
Subtle differences in the individual constructs were considered
here of increased importance. In assessing the semantic similarities
between constructs, two kinds of information were taken into
account. First, qualitative information such as the definition of
each pole for the constructs elicited was used. Secondly, every
construct was characterized by the participants ratings for the set
of stimuli. Quantitative techniques such as Hierarchical Cluster
Analysis provided information related to the cognitive similarity
of the constructs (i.e., how similarly two constructs were being
used in differentiating the items in the set of products). This was
an iterative procedure in which both qualitative and quantitative
information were used to inform the grouping process (see Figure
2).
The hierarchical cluster analysis augmented qualitative
understanding by highlighting: a) constructs that displayed a
high correlation in the ratings, but for which there was no a-priori
identified semantic similarity (from the content analysis), and
b) the cognitive dissimilarity of two constructs that displayed
high semantic similarity. In this sense, for two constructs to
be judged as similar, they not only had to agree with regard to
semantic information, but also with regard to participants ratings
for the set of products. This process was found to provide a rich
qualitative understanding of non-contiguous constructs, in which
the opposite pole doesnt constitute a negation or a linguistic
opposition (Karapanos & Martens, 2008).

60

International Journal of Design Vol.3 No.3 2009

O. Tomico, E. Karapanos, P. Lvy, N. Mizutani, and T. Yamanaka

Table 5 characterizes the descriptive richness of the

different thematic categories. More precisely, it illustrates the
diverse ways in which the Dutch and Japanese designers referred
to the pragmatic and hedonic qualities of the products.
The results show that both the Japanese and Dutch designers
characterized utility by the degree of smoothness experienced
in writing with the pen and with its versatility. However, the
Japanese group added more detail to their characterization by
also considering the thickness of the line produced, as well as
the feeling of balance that the pen provided, and the degree of its
multi-functionality and adjustability. Ease-of-use was described
by both groups in relation to understandability of the functionality
and the mechanisms of the pen. The Dutch designers related
ease-of-use to efficiency and the Japanese to unity of the pens
form and function. Comfort was commonly defined according to
thickness and size of the pen, the softness of the grip and the effort
required for usage, but the Japanese designers added more subtle

elements such as consistency in shape and size, and the fluency

of the pens mechanism and its lightness. Opposite to this, both
groups related durability to the hardness of the material used in the
pens construction, although the Dutch designers also considered
reliability, robustness and breakability as other variables to take
into account.
The Dutch designers placed significantly greater
importance than the Japanese designers on social considerations
related to identification. Both groups considered value to be the
main element driving identification, but the Dutch designers added
more detail to the concept by talking about the pens uniqueness,
its exclusivity, its style and its degree of innovation.
Furthermore, while the Japanese designers referred to the
aesthetics of product appearance more frequently than the Dutch
designers, the Dutch group seemed to be more concerned with the
aesthetics of interaction. In particular, the aesthetics of appearance
were characterized by both groups according to the look of the

Figure 2. Sample of the information used, combining results from the hierarchical cluster analysis and the content analysis.
Table 5. Characterization of the different personal construct thematic categories (categories with higher descriptive richness are
highlighted in bold).
Category

Pragmatic
aspects

Stimulation

Dutch characterization

Japanese characterization

Utility

Versatility; smoothness

Smoothness of writing; thickness of the line;

balance; multifunctionality; adjustability

Durability

Solidness; reliability; robustness;

breakability

Hardness of the material

Ease-of-use

Efficiency; understandability

Understandability of the mechanism; indications; unity

Comfort

Size; grip softness; usage effort

Lightness; shape/size consistency; grip surface;

usage effort; fluency of mechanism; thickness

Unexpected
functionality

Surprise factor; hidden functionality; special

functionality

Fun factor; peculiar sound

Aesthetics of
interaction

Mechanism complexity; feeling of fluency;

invitation to be used

Mechanism complexity

Visual aesthetics

Appearance of material; light reflection;

colorfulness

Shape integration; naturalness; colorfulness;

appearance of material; texture; light reflection;
graphic style; transparency

Exclusivity; specialness; style; value;

innovativeness

Value if it were lost; exclusivity; style

Identification

www.ijdesign.org

61

International Journal of Design Vol.3 No.3 2009

The Repertory Grid Technique as a Method for the Study of Cultural Differences

design relying on global cross-domain psychological constructs.

We have provided an alternative methodological approach to
exploring cross-cultural differences by studying individual
idiosyncratic views of existing designs (using the Repertory Grid
Technique that originated with Kellys (1995) Personal Construct
Theory). We have proposed a content-analytic procedure that aims
at uncovering cultural insight, composed of three measures: a) the
relative percentage of a construct category over the total sample
of constructs, signifying the dominance of a quality for a given
product domain, b) the average elicitation order per construct
category, relating to the subjective importance (i.e., salience) and
implication potential (i.e., importance in preference judgments)
for a given product domain; and c) the descriptive richness of the
categories generated.
These indices can be easily obtained and can be directly
related to any field of design that needs to be assessed, as shown
in the case study. The indices improved the reliability of the
analysis by providing domain-specific and user-generated product
attributes rather than universal items. However, the indices were
not directly calculated from the constructs elicited. They refer
rather to the categories generated by the content analysis, the
reliability of which depends on the expertise of the researchers
conducting the analysis. Future work will focus on applying
different indices developed in Constructivist Psychology to
analyze different Repertory Grids at the same time without any
intermediate analysis.
Moreover, in this case study we explored cultural differences
between Dutch and Japanese designers by identifying the
perceptions of participants with similar professional experience,
similar preferences of use and similar attitudes regarding the
amount of money they were willing to pay for a product. The
designers idiosyncratic views on the set of products, in this sense,
uncovered the qualities that they valued with regard to the existing
product domain through measures of the dominance, importance
and descriptive richness of their personal constructs. From the
results obtained, it is palpable that culture does play a role in those
aspects of a product that designers consider important. However,
the intention of this case study was to exemplify the procedure
proposed and to show that it can generate valid data for analyzing
differences in perception and decision-making processes that can
relate to cultural influences.

material that the pen was made from, and the quality of light
reflection and colorfulness, but, in addition to this, the Japanese
designers focused more on the pens graphic style, its texture and
the integration of shape. Opposite to this, the Dutch designers
added more accuracy in their description of the aesthetics of
interaction by considering important a feeling of fluency in using
the pen and a sense of the pen offering an invitation to be used,
besides the commonly agreed-on aspect of the complexity of the
pens mechanism.

Overview of the Results

Analyzing the results from the three indices (see Table 6) helped
to obtain a general overview and to analyze categories for which
one of the indices had no significant values (cells with one plus
or minus instead of two in Table 6). The descriptive richness is
related to the other two indices, thus adding redundancy to the
analysis. It is important to note that in adding this third index, the
ease-of-use category turned out to be the only category that had
two indices with no significant differences.
The general comparison of the three indices showed that
the Japanese designers were more concerned with the pragmatic
aspects of utility and comfort (both show a higher degree of
dominance and descriptive richness) and the Dutch designers
were more concerned with durability (showing a higher degree
of importance and descriptive richness). Further, while the
Japanese designers referred to the visual aesthetics of the
products more frequently than the Dutch designers (with higher
levels of dominance and descriptive richness), the Dutch group
seemed to be more concerned with the aesthetics of interaction
and unexpected functionalities of the product (with higher levels
of dominance, importance and descriptive richness). Finally, the
Dutch designers were found to pay more attention (with higher
levels of dominance and importance) than the Japanese to the
symbolic qualities of the product, i.e., its ability to communicate
a favorable image of the owner.

Conclusions
In this article we have highlighted a shortcoming inherent in many
current approaches to exploring cross-cultural differences in

Table 6. Overview of the dominance, importance and descriptive richness indices for the Japanese and Dutch designers.
Categories marked with one plus or one minus sign are those in which there were no significant differences between the two groups.
Category

Pragmatic
aspects

Stimulation

Identification

www.ijdesign.org

Dominance

Importance

Dutch

Japanese

Utility

--

++

Durability

++

++

--

Comfort

--

++

Unexpected functionality

++

--

Aesthetics of interaction

++

--

Visual aesthetics

--

++

++

--

Ease-of-use

62

Dutch

Descriptive Richness

Japanese

Dutch

Japanese

--

++

--

++

--

--

++

++

--

++

--

++

--

--

++

++

--

International Journal of Design Vol.3 No.3 2009

O. Tomico, E. Karapanos, P. Lvy, N. Mizutani, and T. Yamanaka

10. Karapanos, E., Hassenzahl, M., & Martens, J. (2008). User

experience over time. In Proceedings of the 26th SIGCHI
Conference on Human Factors in Computing Systems
(extended abstracts, pp. 3561-3566). New York: ACM.
11. Karapanos, E., Martens, J. -B., Hassenzahl, M. (2009).
Accounting for diversity in subjective judgments. In
Proceedings of the 27th International Conference on Human
Factors in Computing Systems (pp. 639-648). New York:
ACM.
12. Karapanos, E., Zimmerman, J., Forlizzi J., & Martens, J.-B.
(2009). User experience over time: An initial framework. In
Proceedings of the 27th International Conference on Human
Factors in Computing Systems (pp. 729-738). New York:
ACM.
13. Kluckhohn, C. (1951). Values and value orientations in the
theory of action. In T. Parsons & E. A. Shils (Eds.), Toward
a general theory of action (pp. 388-433). Cambridge, MA:
Harvard University Press.
14. Kelly, G. (1955). The psychology of personal constructs (Vol.
1-2). London: Routledge.
15. Krippendorff, K. (2004). Content analysis: An introduction
to its methodology. Thousand Oaks, CA: Sage.
16. Lee, K. P., & Harada, A. (2000). Cultural effects on
subjective preference-cross cultural study between Korean
and Japanese. Kansei Engineering International, 1(2), 51-60.
17. Marcus, A. (2000). Cultural dimensions and global web user
interface design: What? So what? Now what? Retrieved
December 10, 2008, from http://www.amanda.com/
resources/hfweb2000/AMA_CultDim.pdf
18. Mcdonagh, D., & Adams-Webber, J. (1987). The implication
potential of personal constructs in relation to their subjective
importance and order of elicitation. Social Behavior and
Personality, 15(1), 81-86.
19. Oshlyansky, L., Cairns, P., & Thimbleby, H. (2006). A
cautionary tale: Hofstedes VSM revisited. In Proceedings
of the 20th BCS HCI Group Conference (Vol.2, pp. 11-15).
London: British HCI Group.
20. Schwartz, S. (1992). Universals in the content and structure
of values: Theoretical advances and empirical tests in 20
countries. Advances in Experimental Social Psychology, 25,
1-65.
21. Tomico, O. (2007). Subjective experience information
gathering and inspiring techniques in interaction design.
Unpublished doctoral dissertation, Technical University of
Catalonia (UPC), Barcelona, Spain.
22. Tong. M. C., & Robertson, K. (2008). Political and cultural
representation in Malaysian websites. International Journal
of Design, 2(2), 67-79.
23. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D.
(2003). User acceptance of information technology: Toward
a unified view. MIS Quarterly, 27(3), 425-478.

To sum up, as many design choices are not likely to be

grounded in empirical data, designers cultural backgrounds will
evidently be carried over in their design preferences and in the
outcomes of their designs. Thus, the three indices presented in
this article could be the starting point of a series of measures that
could be used to analyze quantitatively subjective information
from different participants for a particular product domain. The
development of a procedure that retains the idiosyncratic views
of the participants increases the focus and level of detail of the
analysis compared to previous research that uses questionnaires
with standardized items.

Acknowledgements
This research was supported by the Initiatives for Attractive
Education in Graduate Schools, sponsored by the Ministry of
Education, Culture, Sports, Science and Technology of Japan.

References
1. de Leur, K. R., Drukker, J. W., Christiaans, H. H. C. M.,
& de Rijk, T. R. A. (2006). Cultural differences in product
design: A study of differences between the South Korean and
the Dutch kitchen environment. Journal of Design Research,
5(1), 1633.
2. Feixas, G., & Cornejo-Alvarez, J. M. (2002). A manual
for the repertory grid: Using the GRIDCOR programme
(version 4.0). Retrieved May 28, 2006, from http://www.
terapiacognitiva.net/record/pag/index.htm
3. Fleiss, J. L., Levin, B., & Paik, M. C. (2003) Statistical
methods for rates and proportions. Hoboken, NJ: WileyInterscience.
4. Hassenzahl, M. (2004). The interplay of beauty, goodness,
and usability in interactive products. Human-Computer
Interaction, 19(4), 319-349.
5. Hassenzahl, M., & Wessler, R. (2000). Capturing design
space from a user perspective: The repertory grid technique
revisited. International Journal of Human-Computer
Interaction, 12(3 & 4), 441-459.
6. Hofstede, G. (1984). Cultures consequences: International
differences in work-related values. Newbury Park, CA: Sage.
7. Hsieh, H. F., & Shannon, S. E. (2005). Three approaches to
qualitative content analysis. Qualitative Health Research,
15(9), 1277-1288.
8. Hsu, S. H., Chuang, M. C., & Chang, C. C. (2000). A semantic
differential study of designers and users product form
perception. International Journal of Industrial Ergonomics,
25(4), 375-391.
9. Karapanos, E., & Martens, J. -B. (2007). Characterizing
the diversity in users perceptions. In C. Baranauskas, P.
Palanque, J. Abascal, & S. D. J. Barbosa (Eds.), Proceedings
of the 11th IFIP TC13 International Conference on HumanComputer Interaction (pp. 515-518). Berlin: Springer.

www.ijdesign.org

63

International Journal of Design Vol.3 No.3 2009